Xerox Corporation

Google Inc.

Conduent Business Services, LLC

Wei Hua Peng

Tong Sun

Sudhendu Rai

Saurabh Singh Kataria

Shi Zhao

Sumit Bhatia

Lei Li

Margaret Helen Szymanski

Philip Crane Rose

Shriram V Revankar

Jonas Karlsson

Gabe Culbertson

Julien Jean Lucien Bourdaillet

Nicolas Crowell

Jingxuan Li

Dae Hoon Park

David Thomas Ashby

In one aspect, a method includes obtaining videos and for each video: obtaining a set of anchors for the video, each anchor beginning at the playback time and including anchor text; identifying, from text generated from audio of the video, a set of entities specified in the text, wherein each entity in the set of entities is associated with a times stamp at which the entity is mentioned; determining, by a language model and from the text generated from the audio of the video, an importance value for each entity; for a subset of the videos, receiving rater data that describes, for each anchor, the accuracy of the anchor text in describing subject matter of the video; and training, using the human rater data, the importance values, the text, and the set of entities, an anchor model that predicts an entity label for an anchor for a video.

Video anchors

Systems and methods for controlling access to data in applications using client-side encryption. In that regard, in some examples, a first application (e.g., an email application, calendar application, messaging application, word processing application, file storage application, etc.) hosted from a particular web domain may be configured to invoke a second application hosted from a different origin (e.g., a different web domain or subdomain) to handle receiving and encrypting any sensitive information from a client entered through a client application (e.g., a web browser), and to handle decrypting information to be provided to the client through the client application. This second application may be loaded in an inline frame or similar subwindow or subroutine configured to prevent or limit the first application from having access to sensitive information in the second application.

Systems and methods for controlling data access in client-side encryption

A method, non-transitory computer readable medium, and apparatus for recommending a topic-cohesive and interactive implicit community are disclosed. For example, the method receives a request for customer care, selects an implicit community identified from a plurality of individual users of a social media website based upon a relevance score related to a topic of the request for customer care and recommends the implicit community in response to the request for customer care.

Method and apparatus for recommending topic-cohesive and interactive implicit communities in social customer relationship management

The present disclosure provides a system that allows for the real-time and online monitoring of the exchanges between customers and a CRM team over social media. While crawling all messages exchanged over the social media by customers and CRM team, the system aggregates related messages exchanged between a given customer and the CRM team into a conversation. The system includes a linguistic framework for the analysis of conversations (based on the two linguistic theories of dialog acts and conversation analysis) to label the nature of the messages in a conversation or thread.

Conversation analysis of asynchronous decentralized media

A system and method for analyzing social media data by obtaining social media data from a social media platform, where the social media data includes documents from multiple users of the social media platform; classifying the documents using a sentiment classifier; tokenizing the documents into terms; associating a sentiment with each term; detecting a first event based on a number of occurrences of a first term in the documents; and providing information associated with the event to a user, where the information includes the first term and a sentiment associated with the first term.

Methods and systems for monitoring and analyzing social media data

A method, non-transitory computer readable medium, and apparatus for extracting text from a social media document are disclosed. For example, the method indexes a plurality of social media documents into a plurality of snippets, receives a query including one or more keywords and a purpose, identifies one or more of the plurality of snippets that include the one or more keywords in an index, ranks the one or more of the plurality of snippets in accordance with the purpose and provides the one or more plurality of snippets that are ranked in accordance with the purpose.

Method and apparatus for extracting portions of text from long social media documents

Embodiments generally relate to systems and methods for detecting topics in social media data. More particularly, the systems and methods can extract a concept hierarchy from a set of data, wherein the concept hierarchy comprises a plurality of layers. Further, the systems and methods can train topic models based on the content in each of the layers. Still further, the systems and methods can select the most appropriate topic model for social media data by balancing the complexity of the model and the accuracy of the topic detection result. Moreover, the systems and methods can use the most appropriate topic model to detect topics in social media data.

Systems and methods for scalable topic detection in social media

Embodiments generally relate to systems and methods for generating a sentiment dictionary and calculating sentiment scores of adjectives within the sentiment dictionary. A set of seed words can be identified and expanded using synonyms and antonyms of the set of seed words. Social media data can be parse to identify adjectives that link to the set of seed words with the words 'and' or 'but.' Matrices representing the attraction and repulsion among the linked adjectives can be generated. A factorization algorithm can be minimized to determine an output matrix that comprises positive and negative sentiment scores for each of the adjectives. In embodiments, a sentiment score for part of all of the social media data can be calculated using the output matrix, and one or more parts of the social media data can be classified as a positive or negative sentiment.

Systems, methods and devices for generating an adjective sentiment dictionary for social media sentiment analysis

A method of estimating the cost of a target print job may include identifying a target print job having a document type and one or more attributes, for each attribute of the target print job, determining a correlation between the attribute and a cost of the target print job using a plurality of historical print jobs associated with the document type, and identifying one or more of the attributes as cost drivers based on the correlation of the attribute to the cost of the target print job. The method may include identifying one or more relevant historical print jobs from the plurality of historical print jobs based on values for the identified cost drivers, estimating a cost of the target print job using the one or more relevant historical print jobs, and displaying the estimated cost associated with the target print job.

System and method of estimating the cost of a print job

Embodiments described herein are directed to managing document interactions in a collaborative document area of a virtual world. Document interactions of avatars in the collaborative document area of the virtual world are captured by an interaction tool deployed in the collaborative document area. The document interactions are related to at least one document in the collaborative document area. The document interactions are associated with the at least one document based on a reference scheme applied to the collaborative document area by an interaction association unit.

Managing document interactions in collaborative document environments of virtual worlds

A method of recommending vendors to bid on a print job may include identifying a print job for which a recommendation of vendors to bid on the print job is desired and identifying one or more vendors as potential bidders for the print job. The method may include, for each identified vendor, determining, by a computing device, a bidding probability associated with the vendor, a winning probability associated with the vendor, a recommendation probability associated with the vendor, and identifying the vendor as a recommended vendor based on the associated recommendation probability. The method may include notifying a user of the recommended vendors.

Methods and systems for recommending vendors to submit bids for a print job

Methods and systems for clustering information items using nonnegative tensor factorization are disclosed. A processing device receives one or more class labels, each corresponding to an information item, a selection for a nonnegative tensor factorization model having an associated objective function and one or more parameter values, each corresponding to one of one or more penalty constraints. The processing device determines a constrained objective function based on the objective function associated with the selected nonnegative tensor factorization model, the one or more parameter values and the one or more class labels and including the one or more penalty constraints. The processing device determines clusters for the plurality of information items by evaluating the constrained objective function. Pairwise constraints may be received in addition to or instead of the class labels.

Constrained nonnegative tensor factorization for clustering

A system and method for identifying a key influencer in a social media environment for enterprise marketing utilizing topic modeling and social diffusion analysis. A user interest profile can be generated by analyzing historical data stored in a database utilizing. A social graph can be generated and an influence measuring process based on the social graph data can be performed utilizing a static diffusion model and a dynamic diffusion model to calculate a set of key influencers. The dynamic diffusion model considers time stamp information to assess an impact of each user communication on the growth of a conversation within a time period. The key influencer can be identified in a specific topic area and a number of total users that can be reached via the influencer within a specific time window can be predicted.

Method and system for identifying a key influencer in social media utilizing topic modeling and social diffusion analysis

One exemplary method embodiment, pre-processes customer requests that are maintained in a dataset to create a matrix between products and the customer requests. Each of the customer requests comprises at least a customer identification, a textual request, and a product identification related to the textual request. After such pre-processing of the dataset, the method can respond to queries of the dataset using the matrix.

Semantic enhanced link-based ranking (SEL Rank) methodology for prioritizing customer requests

What is disclosed is a novel system and method for analyzing multi-dimensional cluster data sets to identify clusters of related documents in an electronic document storage system. Digital documents, for which multi-dimensional probabilistic relationships are to be determined, are received and then parsed to identify multi-dimensional count data with at least three dimensions. Multi-dimensional tensors representing the count data and estimated cluster membership probabilities are created. The tensors are then iteratively processed using a first and a complementary second tensor factorization model to refine the cluster definition matrices until a convergence criteria has been satisfied. Likely cluster memberships for the count data are determined based upon the refinements made to the cluster definition matrices by the alternating tensor factorization models. The present method advantageously extends to the field of tensor analysis a combination of Non-negative Matrix Factorization and Probabilistic Latent Semantic Analysis to decompose non-negative data.

Growing community of inventors

San Francisco, CA, United States of America

Wei Hua Peng